[jira] [Commented] (HIVE-17287) HoS can not deal with skewed data group by

[liyunzhang_intel (JIRA)]

    [ 
https://issues.apache.org/jira/browse/HIVE-17287?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16125326#comment-16125326
 ] 

liyunzhang_intel commented on HIVE-17287:
-----------------------------------------

[~lirui]:  in current case, i have not set {{hive.spark.use.groupby.shuffle}}, 
but i think the value is true because the default value is true. After 
disabling {{spark.shuffle.reduceLocality.enabled}}, i reran the query67.  It 
showed passed. But one strange thing i found is [not all stages finished but 
the result of spark job is 
completed|https://issues.apache.org/jira/secure/attachment/12881692/not_stages_completed_but_job_completed.PNG].I
 don't know whether this is bug of spark. Meanwhile in spark history server, 
the shuffle read metrics are still skewed. 

> HoS can not deal with skewed data group by
> ------------------------------------------
>
>                 Key: HIVE-17287
>                 URL: https://issues.apache.org/jira/browse/HIVE-17287
>             Project: Hive
>          Issue Type: Bug
>            Reporter: liyunzhang_intel
>            Assignee: liyunzhang_intel
>         Attachments: not_stages_completed_but_job_completed.PNG, 
> query67-fail-at-groupby.png, query67-groupby_shuffle_metric.png
>
>
> In 
> [tpcds/query67.sql|https://github.com/kellyzly/hive-testbench/blob/hive14/sample-queries-tpcds/query67.sql],
>  fact table {{store_sales}} joins with small tables {{date_dim}}, 
> {{item}},{{store}}. After join, groupby the intermediate data.
> Here the data of {{store_sales}} on 3TB tpcds is skewed:  there are 1824 
> partitions. The biggest partition is 25.7G and others are 715M.
> {code}
> hadoop fs -du -h 
> /user/hive/warehouse/tpcds_bin_partitioned_parquet_3000.db/store_sales
> ....
> 715.0 M  
> /user/hive/warehouse/tpcds_bin_partitioned_parquet_3000.db/store_sales/ss_sold_date_sk=2452639
> 713.9 M  
> /user/hive/warehouse/tpcds_bin_partitioned_parquet_3000.db/store_sales/ss_sold_date_sk=2452640
> 714.1 M  
> /user/hive/warehouse/tpcds_bin_partitioned_parquet_3000.db/store_sales/ss_sold_date_sk=2452641
> 712.9 M  
> /user/hive/warehouse/tpcds_bin_partitioned_parquet_3000.db/store_sales/ss_sold_date_sk=2452642
> 25.7 G   
> /user/hive/warehouse/tpcds_bin_partitioned_parquet_3000.db/store_sales/ss_sold_date_sk=__HIVE_DEFAULT_PARTITION__
> {code}
> The skewed table {{store_sales}} caused the failed job. Is there any way to 
> solve the groupby problem of skewed table?  I tried to enable 
> {{hive.groupby.skewindata}} to first divide the data more evenly then start 
> do group by. But the job still hangs. 



--
This message was sent by Atlassian JIRA
(v6.4.14#64029)